Edit Existing PDFs

Use parseDocument when you need to inspect a PDF and editDocument when you need to produce an incremental update.

import { editDocument, parseDocument } from "@criston/zeropdf";

const parsed = parseDocument(existingBytes);

console.log(parsed.version);
console.log(parsed.listPages().length);
console.log(parsed.listFormFields());

const editable = editDocument(existingBytes);
editable.updateInfo({ title: "Updated document" });
editable.setFieldValue("status", "approved");

const updatedBytes = editable.toUint8Array();

Page workflows

The page transfer helpers cover common document assembly tasks:

import {
  appendPages,
  extractPages,
  mergeDocuments,
  splitDocument
} from "@criston/zeropdf";

const firstPageOnly = extractPages(sourceBytes, [0]);
const combined = mergeDocuments([coverBytes, bodyBytes, appendixBytes]);
const perPage = splitDocument(sourceBytes); // one single-page PDF per page
const withAppendix = appendPages(sourceBytes, appendixBytes);

Encrypted PDFs

Generated PDFs can be password protected, and parser/editing workflows can accept a password when reading protected input.

import { createDocument, parseDocument } from "@criston/zeropdf";

const protectedDoc = createDocument({
  encryption: {
    algorithm: "rc4-128",
    userPassword: "reader",
    ownerPassword: "owner-secret",
    permissions: {
      printing: false,
      modifying: false,
      copying: false,
      annotating: false
    }
  }
});

const bytes = protectedDoc.toUint8Array();
const parsed = parseDocument(bytes, { password: "reader" });

Content Stream Overlay

writePageContent() lets you inject raw content stream operators onto an existing page without altering the original objects. The mode controls placement:

const editable = editDocument(sourceBytes);

editable.writePageContent(0, "1 0 0 rg 72 72 100 100 re f", "append");
editable.writePageContent(0, "q BT /F1 12 Tf 72 700 Td (Overlay) Tj ET Q", "prepend");

Mode	Behavior
"append"	Add operators after the existing page content (default)
"prepend"	Insert operators before the existing page content
"replace"	Replace the entire content stream with the provided content

Xref Recovery

When parseDocument encounters a damaged or missing cross-reference table, it automatically attempts reconstruction by scanning the file for object markers. This handles cases where the startxref pointer is invalid or xref entries point to wrong offsets.

No additional API call is needed — the recovery runs transparently during parseDocument().

Repair Warnings

Call getRepairWarnings() after parsing to inspect what the xref recovery fixed:

const parsed = parseDocument(bytes);
const warnings = parsed.getRepairWarnings();
for (const w of warnings) {
  console.log(`[${w.objectNumber}] ${w.message}`);
}

Each warning includes the affected object number and a description of the repair.

Encrypted PDF without Permissions

If an encrypted PDF omits the permissions (/P) entry, the parser defaults to granting all permissions (print, modify, copy, annotate, etc.). This ensures the document is editable without needing an owner password.